Before this experience, my leadership style was mainly focused around being a good example. The way I look at leadership is that to lead a group of people it’s best to lead by example, along with clear communication on how to do certain things. Whether that’s a group project, or on a team, I like showing what needs to be done rather than just telling. That stems from my first job at my Brazilian jiu-jitsu gym in high school, where my boss Dan sat down with me every day of my first week, showing me how to sell merchandise, create new members in their management system, and take care of the facility. To me that felt like I was being shown the right way to do things, and I’ve carried that leadership style with me for a long time. Because of that, I learned that when I’m being led, I thrive when given clean, consistent instructions and expectations. However, one thing I’m realizing is that clear instructions can kill creativity. In my internship, the head of the data team, Tomas, has a leadership style I haven’t really encountered before. When I had a check-in meeting with him after my first week, he proposed this research project to me, where the simple goal was just to research one person’s idea for an LLM wiki. Two days later I came back to him with my findings on how it works and my thoughts, and he told me that he wants me to think bigger on how to apply such an idea to the data team and then the whole company. Basically, the gist of this project is about giving autonomous AI the right context to complete a task. He specifically didn’t tell me that other people in the data team had attempted this, and he wanted me to just figure it out on my own. Looking back now, it’s clear that he was running an experiment to see if I would come up with a completely different solution than the past iterations of this project. And I relatively did — what I was able to contribute was a real architecture to their existing design, which was extremely messy. After two weeks of research, I made a PowerPoint and presented it to Tomas and the two other people who had tried this project. After meeting with the two team members who had tried it, I had to re-explain my theory to them, and I was able to cleanly explain everything I had come up with based on my research. They were simply confused about the difference between their repository and the structure of mine. After multiple follow-up meetings, we connected the dots that I hadn’t even seen their repository. Once I saw their work, I could see how my architecture could greatly benefit an already functional system. Since then, we’ve started working on applying the theory to their current living repository.

Reflecting on the process of the LLM wiki, I’m honestly happy that I wasn’t given too much direction. I think that if I had been given explicit instructions rather than just being told to open-endedly research this idea, I wouldn’t have come up with such a strong theory. To me, that really helped me be creative and look at this project from as many different angles as possible. Overall, I think that now my idea of leadership looks a lot different. It’s more than just showing people how things are done and telling them to do so. It’s more than that — it’s about inspiring, and sometimes removing restrictions to let someone freely come up with a solution on their own.